English

Minimax Optimal Algorithms with Fixed-$k$-Nearest Neighbors

Statistics Theory 2024-09-11 v3 Distributed, Parallel, and Cluster Computing Information Theory Machine Learning math.IT Machine Learning Statistics Theory

Abstract

This paper presents how to perform minimax optimal classification, regression, and density estimation based on fixed-kk nearest neighbor (NN) searches. We consider a distributed learning scenario, in which a massive dataset is split into smaller groups, where the kk-NNs are found for a query point with respect to each subset of data. We propose \emph{optimal} rules to aggregate the fixed-kk-NN information for classification, regression, and density estimation that achieve minimax optimal rates for the respective problems. We show that the distributed algorithm with a fixed kk over a sufficiently large number of groups attains a minimax optimal error rate up to a multiplicative logarithmic factor under some regularity conditions. Roughly speaking, distributed kk-NN rules with MM groups has a performance comparable to the standard Θ(kM)\Theta(kM)-NN rules even for fixed kk.

Keywords

Cite

@article{arxiv.2202.02464,
  title  = {Minimax Optimal Algorithms with Fixed-$k$-Nearest Neighbors},
  author = {J. Jon Ryu and Young-Han Kim},
  journal= {arXiv preprint arXiv:2202.02464},
  year   = {2024}
}

Comments

65 pages, 5 figures. The manuscript has been revised from scratch compared to the previous version. Notable differences include (1) updated statements and corrected proofs for classification and regression, (2) explicit statements and proofs for distance-selective rules, and (3) new analogous estimators for density estimation

R2 v1 2026-06-24T09:21:19.624Z